At a glance
Access: Images 2.6 is available globally today in ChatGPT and Codex across all subscription tiers on desktop, mobile, and web.
Official announcement: Images 2.6
What shipped
Images 2.6 is OpenAI's latest image release, announced Nov 04 2026 for ChatGPT, Codex and the API. OpenAI says it generates images 50% faster than before, adds GIF output, and gives developers SVG exports, image markup and a new complexity parameter. For creative teams, the headline is shorter edit rounds and steadier characters, products and brand details.
OpenAI introduced Images 2.6 in ChatGPT, Codex and the API. The company's announcement says Images 2.6 generates images 50% faster than before.
In the API, developers can choose between two models: GPT‑Image‑2.6 Flare and GPT‑Image‑2.6 Sunburst.
OpenAI also reports that people now create 4 billion images per week across ChatGPT, Codex and the API. It says that averages more than half a billion images per day.
Key capabilities
Flare for speed. OpenAI says Flare can generate images in as little as six seconds when quality is set to low. It also improves layout quality for posters, product imagery and other professional visuals.
Sunburst for precision. At medium or high quality, OpenAI says Sunburst brings higher-quality generation and greater image consistency across edits.
GIFs. You can animate a brand mascot or character, or turn a product shot into a loop. OpenAI also names reactions and memes for social media, and animated storyboards for exploring character movement before video production.
Consistent edits. OpenAI describes keeping characters, products and brand details consistent as you edit.
API additions. Image markup, already available in ChatGPT, comes to the API so teams can pinpoint where to make specific changes. OpenAI says SVG exports make designs ready to use across websites, apps and print.
Complexity parameter. A new parameter lets developers adjust how much planning the model does for a request. Image quality controls visual detail and fidelity, while complexity guides how much goes into the scene.
Official samples and early work
No official samples yet.
What it means for creative work
Shorter review rounds. The 50% speed figure is OpenAI's own, and BMI Studios has not tested it. Check it against your own briefs before you plan around it.
Consistency is the real test. Brand work depends on the same character, product and details surviving every edit. BMI Studios holds that "Consistency is the hard part, so we build for it." See how that works in Creative Systems & Asset DNA.
Motion before video. OpenAI points to animated storyboards for exploring character movement before video production. That makes GIFs an early step ahead of a full AI video build.
Settings become a creative decision. Picking a model, a quality level and a complexity level per task is now part of the brief. OpenAI has not published how complex a graphic can be before an SVG export falls short.
Tools fit into workflows. Picsart says in OpenAI's post that the 2.6 model brings more accurate results, better character consistency and faster generations. As BMI Studios puts it, "Craft sets the ceiling, not the tool."
How to try it
ChatGPT and Codex. Images 2.6 is available globally today across all subscription tiers on desktop, mobile and web.
API. GPT‑Image‑2.6 Sunburst and GPT‑Image‑2.6 Flare are available in the API.
Editing. OpenAI recommends low complexity for most editing tasks, paired with the image quality your workflow requires. For scenes with many elements or constraints, it suggests trying medium or high complexity.
Drafts and finals. A reasonable split is Flare at low quality for fast drafts and Sunburst at medium or high quality for detail-critical work. That split is our reading of OpenAI's guidance, not an OpenAI recommendation.
Pricing and safety. Pricing details and the system card are linked from OpenAI's announcement.
Planning a workflow around it? Contact BMI Studios.
Sources
- https://openai.com/index/introducing-images-2-6/, accessed Nov 05 2026